Effect of Diffusion and Biological Oxidation inside the floc matrix in the

This experiment was designed to evaluate the effect of biological flocculation and internal diffusion in bacterial flocs in the kinetics of dissolved substrate consumption. Similar batch experiment were run in parallel using two different bacterial suspensions, the first one under normal conditions, and the second one with a deflocculating agent. A DOWEX 50 x 8, 20 - 50 mesh cation exchange resin (CER) was used as a deflocculating agent; this CER removes divalent cations from the sludge matrix destroying the biochemical bridges used by the EPS to flocculate the suspended and colloidal particles which became free in the suspension (Frolund et al. 1996).

In a similar study to the one proposed herein La Motta et al. (2003) found that flocculation itself does not have an important effect in the removal of DCOD, they found no difference between a flocculated and a deflocculated reactor when evaluating the kinetic of a dissolved substrate. On the other hand, Logan and Hunt (1988)argued that since bioflocculation

said that for a culture of microorganism to bioflocculate when substrate is nearly depleted implies that the cell associations may confer some advantage over freely dispersed cells relative to increasing substrate uptake by aggregated cells. Based on a theoretical analysis Logan and Hunt found that bioflocculation increases the rate of substrate transport to cells in permeable flocs compared to dispersed cells. However, they found that the permeability of the floc may define the direction of this relationship between flocculated and dispersed cells. Steiner et al. (1976) expressed that activated sludge flocs remove both colloidal matter and soluble BOD by adsorption. La Motta et al. (2003, 2004) clearly demonstrated the role of bioflocculation in the removal of particulate COD, and the polymers bridging theory supports the idea that bioflocculation is a mechanism for trapping organic particulates previously to hydrolysis. As expressed by Okutman et al. (2001) the rate of hydrolysis is much slower than that of the utilization of soluble readily biodegradable substrate. They results presented in the previous Section indicate that dissolved substrate is still present in the solution after 20 and 30 minutes of residence time when relatively high initial substrate concentration was used; moreover, La Motta et al. (2003, 2004) found remaining DCOD after 20 and 30 minutes of HRT in batch and continuous flow reactors when evaluating artificial and municipal wastewater, but at those times they found a good removal of particulate COD by bioflocculation. They point that the author is trying to answer is why the biomass produced polymers and promoted the aggregation by bioflocculation when there was still soluble COD in the suspension? As expressed by Logan and Hunt (1988): Is there any advantage of growth within an aggregate regarding the consumption of readily biodegradable substrate? The experiments presented herein were designed with the goal of providing an answer to these questions.

Table 4.5 presents the data gathered for the suspension with and without CER. For this experiment flocculated sludge was collected at the aerobic contact chamber and brought to the environmental lab within 40 minutes. Once at the laboratory the samples were analyzed following the procedures describe in Chapter 3. The working solution was prepared with the fresh sludge and artificial wastewater prepared with methanol as a single carbon source.

The mixture of sludge and substrate had an initial concentration of 1250 mg VSS/L and the COD initial value (time 0) was 344 mg DCOD /L yielding an initial So/Xo relationship equal to 0.28. The mixture with the CER, which was added according to the dosification proposed by Frolund et al. (1996), was agitated with a magnetic stirrer at velocity of 50 rpm for five minutes. After agitation the DCOD was again measured. Interestingly, the result for the initial DCOD for the deflocculated sample was 20 mg/L higher after the agitation process; this value is reported in Table 4.5 as So for the deflocculated sample. The increase in the DCOD may be due to the release of EPS and soluble microbial products (SMP) into the suspension after flock break up.

Table 4.5 DCOD at different HRT for flocculated and deflocculated samples for a Medium So/Xo ratio

Unit Time _(min) So Flocculated _{(DCOD, mg/L)} Se Flocculated _{(DCOD, mg/L)} So Deflocculated _{(DCOD, mg/L)} Se Deflocculated _{(DCOD, mg/L)}

1 0 344 344 364 364

2 5 344 123 364 104

3 10 344 40 364 38

4 15 344 20 364 15

Figure 4.19 shows the effect of the residence time in the supernatant DCOD for both the flocculated and the deflocculated sample. The statistical software Data FitTM was used to fit the first-order model with an asymptotic non-biodegradable COD fraction, in both cases excellent R2, higher that 0.99, were obtained. This first-order model represented by Equations 4.6 and 4.7 was selected because it accurately predicted the data for the cases with medium and high So/Xo as the one presented in Table 4.5, and also for its simplicity. Equation 4.7 is used instead of Equation 4.5 because, as shown in Section 4.1 the biomass growth is negligible under medium and high So/Xo ratios.

For the flocculated sample the values obtained with the curve-fitting analysis are: R2 = 0.999 D a = 7.4 mg/L = Dx K 0.22 min-1.

For the deflocculated sample the values obtained with the curve-fitting analysis are: R2 = 0.995 D a = 14.1 mg/L = Dx K 0.28 min-1.

0 50 100 150 200 250 300 350 400 0 10 20 30 Time (min) Se ( DCOD,m g /L )

Flocculated Fit Equation-Flocculated

Deflocculated Fit Equation - Deflocculated

Figure 4.19 Effect of Bioflocculation on the Kinetics of DCOD Consumption for a Medium So/Xo ratio.

Both the first-order kinetic constant,K_Dx, and the asymptotic non-biodegradable COD coefficient, a_D, were higher for the deflocculated sample. The fact that the kinetic constant has a higher value is an indication that the dissolved COD consumption in the deflocculated sample, where there are freely dispersed cells, proceeds at a faster rate; this is and indication that bioflocculation results in a slight limitation in the amount of substrate that is supplied to the bacteria inside the flocs, which reduces the biodegradation rate. On the other hand, the fact that the asymptotic non-biodegradable COD coefficient is also higher for the deflocculated sample looks like a contradiction to the previous statement. However, if EPS were release as soluble substance after the CER was added, the high value of a_D would be an indication of low

Pavoni et al. (1973) on the biodegradability of exocellular polymer substance, where they found a very low level of biodegradability for EPS. The high value of a_D may also be due to the release of SMP during the floc break up during the agitation of the sample with the CER. According to Gaudy and Blachly (1985) over 90% of the residual COD measured in batch reactor is subject to biological degradation; however, this residual COD experimented lower conversion rates than the original substrate used in its generation.

The experiment presented in the previous paragraphs was repeated using a lower So/Xo ratio. The new experiment was carried out following a similar procedure to the one described in the previous paragraphs, and in Chapter 3. The mixture of sludge and substrate had an initial concentration of 2300 mg VSS/L and the COD initial value (time 0) was 220 mg DCOD/L giving a So/Xo value equal to 0.10. After the agitation with CER the deflocculated sample presented a raise in the DCOD; triplicate values indicated that the DCOD was equal to 267 mg/L, 40 mg/L more than the value measured previous to agitation. The increase in the DCOD may be attribute to the release of EPS and SMP products into the suspension after flock break up. Table 4.6 presents the data gathered for the suspension with and without the cation exchange resin.

Table 4.6 DCOD at different HRT for flocculated and deflocculated sample for a Low So/Xo ratio

Unit _(min) Time So Flocculated _{(DCOD, mg/L)} Se Flocculated _{(DCOD, mg/L)} So Deflocculated _{(DCOD, mg/L)} Se Deflocculated _{(DCOD, mg/L)}

1 0 220 220 267 267

2 5 220 113 267 90

3 10 220 60 267 62

4 15 220 42 267 35

Figure 4.20 presents the kinetics of the DCOD for both cases, i.e, the flocculated and the deflocculated samples for the low So/Xo ratio.

0 50 100 150 200 250 300 0 5 10 15 20 25 30 Time (min) S e (DCO D, m g /L )

Flocculated Fit Equation-Flocculated

Deflocculated Fit Equation - Deflocculated

Figure4.20 Effect of Bioflocculation on the Kinetics of DCOD Consumption for a Low So/Xo ratio.

Equation 4.7 was fitted to the data using the software Data FitTM, with very good correlation in both cases. The results of the fitting analysis are presented in Table 4.7

Table 4.7 First-order Kinetic Constant for Flocculated and Deflocculated Samples for a low So/Xo ratio. Flocculated Deflocculated R2 0.987 0.989 D a (mg/L) _{2 20} K , min-1 0.14 0.23

Like in the case for the medium So/Xo ratio, both the first-order kinetic

constant,K_Dx, and the asymptotic non-biodegradable COD coefficient, a_D, were higher for the deflocculated sample, supporting the discussion presented previously.

Table 4.8 presents a summary of the first-order kinetic constants obtained with the flocculated and deflocculated samples.

Table 4.8 First-order Kinetic Constant.

Flocculated Deflocculated So/Xo 0.28 0.08 0.1 0.28 0.1 Dx K =K_Dx.X (min-1₎ 0.22 0.33 0.135 0.28 0.23 X (mg/L) 1300 3000 2292 1300 2292 D

K (L/mgVSS min) _{1.7E-04 1.1E-04 5.9E-05 2.2E-04 1.0E-04}

Figure 4.21 shows the relationship between the first-order kinetic constant K_D and the initial soluble substrate to biomass ratio. Even though the relationship is not completely well defined it can be said that the value of the kinetic constant tends to increase as So/Xo increases.

R2 = 0.7983 0 0.00005 0.0001 0.00015 0.0002 0.00025 0 0.05 0.1 0.15 0.2 0.25 0.3 So/Xo (dimensionless) K d ( L/M gV S S m in) Flocculated Deflocculated

Figure 4.21 First-Order Kinetic Constant K_D versus So/Xo

Figure 4.22 shows the values obtain for KDx = KD.X for different So/Xo ratios. It can be

concluded from this Figure that the value of K_Dx is independent from the initial soluble substrate to biomass ratio.

R2 _{= 0.0051} 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0 0.05 0.1 0.15 0.2 0.25 0.3

So/Xo (dimensionless)

Kd

.X

( m

in

)

CHAPTER V

5.1 CONCLUSIONS

The main purpose of this investigation was to study the kinetics of readily biodegradable soluble substrate simulated with methanol as a single carbon source and measured as dissolved chemical oxygen demand. The kinetics of soluble substrates was evaluated with different models and at different So/Xo ratios, and also using flocculated and dispersed cells suspensions. The specific conclusions that can be drawn from this research are:

• The removal of readily biodegradable soluble substrate at medium and low So/Xo rations (values less than 0.3) can be well described by a first-order kinetics with an asymptotic non-biodegradable portion as the one presented next:

X a S K

r_su =− _D( _D − _D)

Values of the product K_DX varied between 0.14 and 0.33 min-1. Most of the municipal wastewater presents a ratio So/Xo less than 0.3, and therefore the first- order rate expression presented herein for the removal of readily biodegradable soluble substrate can be used in conjunction with bioflocculation kinetics or hydrolysis kinetics for the simulation of complex dissolved-particulate substrates.

• For values of So/Xo less than 0.3 the growth of biomass is negligible, and thus can be neglected in the determination of the kinetics of soluble substrates.

• A directly proportional relationship was found between the yield coefficient, Y, and the So/Xo ratio. This is an indication that at low So/Xo ratio the cells are basically using the substrate for maintenance and not for growth. This is a contradiction of the classical Monod Equation which does not consider the fact that microorganism may need substrate even when they do not grow. As the So/Xo ratio increases the biomass uses more substrate for the synthesis of new cellular material.

• The value of the first-order kinetic constant K_D tends to increase as the value of the initial soluble substrate to biomass increases. However, the value of the product of

D

K times the biomass concentration is independent of the So/Xo ratio.

• First-order kinetics can describe very well the consumption of readily

biodegradable soluble substrate for both freely dispersed cells, and flocculated suspensions.

• The dissolved COD consumption for freely dispersed cells proceeds at a faster rate than for flocculated suspensions. This is an indication that the diffusion in the flocs results in a limitation in the amount of substrate that is supply to the bacterias that are inside the floc.

• The value of the first-order kinetic constant K_D tends to increase as the value of the initial soluble substrate to biomass increases. However, the value of the product of

D

K times the biomass concentration is independent of the So/Xo ratio.